It seems to me that's a solution looking for a problem. Don't like moderation? Read at -1. It's hardly a tyranny that the posts you most want to read aren't always at the top of the list. The typical (non-MS v. DOJ) story at Slashdot never has so many posts that it would be impractical to read them all if you're worried about missing a gem that wasn't moderated upward because it was posted too late or went against conventional wisdom (personally, I don't see the latter--you can generally find both sides of even contentious issues modded up). This is especially true since most of the lower ranked posts seem to be a single sentence.
If you stop worrying about the popularity contest of moderation, there are likely better ways to find interesting comments than an elaborate system of granting trust to editors. For instance, how about a button next to each post to grant a +/- 1 to other posts by the same author in your view. Using editors as a proxy for your tastes seems oddly indirect for something that you can filter yourself.
Part of the quid-pro-quo of the patent system (temporary monopoly in exchange for juicy details of invention) is that there has to be details of the invention that are non-trivial to implement/discover. If there aren't, then the public is giving up something naturally theirs (the right to use any idea) in exchange for nothing.
This is why according to law, the invention has to not only be original, but also must not be obvious to someone "skilled in the art"--i.e. who knows the technology in the area the patent covers. If it can be trivially reproduced (e.g. from a one-sentence description of what it does), then it was obvious.
The Amazon "one-click" patent ought not to have been awarded, even if it turns out that it was original, because once it is described, any web developer worth her salt can produce the equivalent in a few minutes work.
Because of the "obvious" nature of the idea, there's no public policy reason to grant Amazon a monopoly.
Patents (at least in the US) are not awarded as a gold star for having come up with an original idea, they are awarded to further the public good.
The framers of the Constitution recognized that nobody has or ought to have ownership of an idea, but in order to encourage people to share ideas instead of keeping them secret they made the patent system where you were granted a limited temporary monopoly on the use of your idea--in exchange for which you had to publish your idea, with all the detail necessary for someone else to implement it. Alternatively, you are perfectly within your rights to try to keep your idea secret, but if someone else comes up with the same idea, or your secret leaks out, you have no legal recourse.
So what about the one-click patent? Regardless of whether someone thought of it before, there's no public good in protecting an idea the very description of which gives anyone "skilled in the art" enough to go on to reproduce the idea. It is impossible for Amazon both to use this idea and to keep it a secret, therefor there's no reason to bargain with them to make it public.
The record company cuts her a check as an "advance on royalties"
Against this they charge her:
the cost of recording the album
the cost of pressing and distributing the album
the cost of promoting the album
pretty much anything else they can think of
If, after all their creative accounting, there is any further "debt", they charge that against the next album as well. If the record makes money, the studio ends up not actually paying anything toward the production and distribution--it all comes out of the artist's share.
It's not only theoretically possible for the artist to owe the studio money after a profitable album, it happens all the time. Records go platinum, and the artist goes bankrupt.
So, yeah, Universal ought to owe all their artists a proportional chunk of that money, to show up on their royalty statements.
Discounts for Seniors, members of AAA, Diners Club, etc.
Frequent buyer programs
coupons
Happy Hour
Year-end clearance sales
N-th customer promotions
Buy two, get one free
Nobody has any reasonable expectation that the price they pay will be the same from moment-to-moment regardless of the circumstances under which they buy. Nobody claims there's a conspiracy when two gas stations in the same franchise several blocks from eachother charge different prices, or different franchises across the street from one another. (In fact, quite the opposite: people charge price fixing when there's too much uniformity of price.) That's precisely the reason that some retailers offer a price guarantee as a selling point ("If we lower the price within the next 30 days, we'll refund the difference!").
If the price changes were separated by days, or were explicitly linked to things like which physical store they went into, or membership in a particular organization, nobody would even blink. It's only because the changes were happening at online speeds with no particular promotional effort that anyone thinks otherwise.
Suppose you're on top of the Empire State Building and want to get down to the sidewalk. You could either jump off (faster), or take the elevator down. They both have only two outcomes: You live or you die. Which do you choose?
If you chose the elevator, good for you: you understand why having two outcomes is not the same as having a 50-50 chance.
If you chose to jump, well, to quote Niven and Pournelle, "Think of it as evolution in action..."
A) You're assuming that Shasta and Pepsi are otherwise the same product, so that the only reason there is a difference in sales is advertising. Coca-Cola decided to put that theory to the acid test when they introduced "New Coke", reasoning that if they had a product that tasted just like Pepsi, they could slaughter them with advertising dollars. Didn't work so well, did it?
B) You're assuming that banner ads are just as effective as all other forms of advertising. That's hotly disputed.
C) As far as Double-Click making money--Double-Click is selling ads, not products. It's irrelevant whether Double-Click ads really work as long as they can get enough suckers who think that they work. And the dot-com economy proves there's no shortage of suckers willing to throw money at e-commerce.
Here's a pop-quiz for you: Name an e-commerce company that's making money selling to consumers (no infrastructure companies like Cisco). Now, if you've come up with one, describe one of their banner ads.
Have you ever noticed that whenever someone ends a sentence with ", thank you" (as in "I don't need an electronic conscience in my mail/new reader. My own will do just fine, thank you."), they never mean it? Same with sentences that begin with "I'll thank you to..."
Maybe someone could write a program that would filter for that...
I'd be a lot more concerned about all these attempts to capture "eyeballs" if there was any evidence at all that the eyeballs actually resulted in enough sales to make it worth the advertiser's while. As it is, I fully expect all of these schemes to collapse under the weight of massive consumer indifference.
Mozilla sent them, but Slashdot ignored them. If you want to control the formatting of your text on Slashdot, you need to learn to use the allowed HTML tags.
According to your thesis, this is your fault for not learning the nitty-gritty of the Slashdot interface.
I have nothing against the GPL, but consider what would have to happen before this bone of contention (re: whether the choice of law provision of CNRI's license is GPL compatible or not)would even be an issue.
Suppose you were to use Python 1.6+ under the CNRI license. Under what conditions could you be sued for mixing it with GPL code? As near as I can figure, you would have to have taken the Python code-base, extended it or embedded it within an application that you distribute AND have embedded someone else's GPL'ed code within your application. Then and only then you might be running the risk that the copyright holder of that other piece of code could try to sue you for having violated the terms of the GPL by including his code in a project that wasn't completely GPL compatible. (And if he did two out of the three sets of lawyers that have looked at both licenses agree that he wouldn't have a case.)
To me, this is a ridiculous thing to worry about...you're more likely to be struck by lightning or killed by terrorists than to become the test case where the alleged incompatibility of the GPL and the CNRI license is fought out in court.
The program does automatically detect which open you're talking about; it's just that with "from os import *" you've commanded the program to prefer the os version to the builtin version whereever there's a namespace collision. That's why experienced Python programmers will recommend (I even think this is in the tutorial) that you don't use the "from spam import *" version of the command unless you are importing a module that was specifically designed (e.g. Tkinter) to be used that way.
Basically, Python tries to operate on the principle of least surprise. If you tell it to replace all the local function with the os versions of them, and then invoke the os open function, you shouldn't be too surprised if it behaves the way the native os open function does. You were surprised (this time), but you probably won't run into that again; consider how surprising it would be if it worked the other way around, though: despite explicitly telling the program to always prefer the os versions if available, the program persisted in secretly invoking the builtin version instead?
When you import another module, you do have access to that modules variables--they just don't pollute your local namespace unless you explicitly command them to (with "from spam import eggs" version of the import command). All you have to do is refer to them by their proper name: spam.eggs instead of just eggs.
If you really want to use eggs instead you can:
from spam import eggs
from spam import *
(sucks in all the names from spam)
import spam
eggs = spam.eggs
I don't know what you mean by "the writing function doesn't allow you to write strings." You can write pretty much anything you want, including binary.
Python has if/elif/else, for item in sequence, and while expression as forms of flow control. What else do you need? Endif and Wend are superfluous--they just signal the end of the block (Perl and C use }). Python can tell by the level of indentation.
GOTO. Don't need it. Between break, continue, and exception handling all the legitimate uses of GOTO as a flow control are covered (I suppose labelled breaks might be nice, but I haven't really missed them).
I don't understand your complaint about importing other modules as a form of flow control at all. I think you must have just missed the whole idea of functions and classes. What version of BASIC were you using before? Even VB doesn't use simple chaining of source-files for flow control--at least, not anymore.
The comp.lang.python newsgroup is bar far the friendliest, most informative, most newbie-tolerant newsgroup I've encountered on USENET. If that be hate-based, let's have more of it.
Python has excellent XML support. Check out the newly released book XML Processing with Python(just the first link I found, not a particular endorsement of the site).
Actually, it's a long standing joke in comp.lang.python that Guido van Rossum, the creator of Python, has a time machine. It usually come up in the context of someone saying, "Boy, I really wish Python would add feature X." Then someone (usually Python guru Tim Peters) would respond to the effect that Guido likes the idea of adding feature X so much that he will use his time machine to retroactively add it to the language as of three releases ago....
"I am curious how one would access the same data in a Python construct - ie, how would a python programmer retrieve *all elements of an array pointed to by a an associative array reference* ? "
That's easy. Everything in Python is really a reference to an object, so in Python terms you're talking about an entry in a dictionary (Pythonese for an associative array) that happens itself to be an array object. Ergo:
It's simple, really. The GPL grants people who agree to abide by it more rights to redistribute the software and source code than they would enjoy under regular copyright law. The RIAA is trying to push an interpretation of legislation that would take away your existing rights under copyright law (despite the fact that the law, even the dread DMCA, has built into it explicit provisions for fair use).
The argument against internet regulation remains the same. If installing extra hardware on every network would provide real security, then I think Bruce Schneier would endorse it. Rather, I believe the point not that we are not yet capable of a secure technology (whether implemented in hardware or software), but that there is no conceivable technology that is truly secure. The best you can do is realistically assess risks and mitigate them where it's cost-effective to do so.
You say it will apply to any application that has sufficient integration, I say it will apply to any application that has insufficient security. It would take nothing in particular by way of design to have the word processor default to not trying to pull down non-local images when you open a local document, and to not set cookies. Or even to ask you when the document tries to do so.
As a Linux user, I'm certainly never going to install, let alone use, an integrated application that's designed with such a cavalier attitude towards security. Non-local communication should be the exception, not the norm, and should never occur invisibly.
...in that when you open a word processing document, you don't usually think that you're "doing something on the Internet." You especially don't think that you are communicating and setting cookies via invisible images.
Whoops. Now there's a suggestion that misses the Cluetrain.
"Thank you for your recent message to customer support. Please be advised that there will be a seven to ten day delay while our legal department reviews our customer support staff's reply. Thank you for your patience while we reduce our legal liability."
Find a sympathetic tenured professor and have him/her host whatever it was that attracted the MPAA letter. Academic institutions have completely different standards for freedom of speech for the faculty than the students. If nothing else, it should cause an entertaining ruckus.
Yeah, yeah, freedom means nothing to a starving man. We've heard it before, and people have starved themselves to death proving it wrong.
Aside from the whole unproductive "How can you care about X when Y is so much more important?" whine (That's what it means to be free; deal with it.), you ignore the fact that modern technology--software above all--is the greatest threat to freedom of any kind that the world has ever seen. We are now capable of monitoring populations to a level of detail that exceed the wildest dreams of history's most oppresive regimes. We are rapidly approaching a point where surveillance can go from being total against a targeted individual to ubiquitous against all individuals. What price freedom then?
It is because freedom is more important than software that software is going to be one of the most important battlegrounds in future wars for freedom. Without free (as in speech and beer) software, that battle will be lost. (Heck, you won't even be able to tell that the battle has been engaged.)
Lest this sound like mere apocalyptic ranting, consider the FBI's Carnivore program. How much are you willing to trust the FBI that it will never be used against anyone but lawfully designated targets? Without full disclosure of the source, how much are you willing to trust that it can't be compromised by someone other than the FBI? How would your answer change if the authority providing the assurances were The People's Republic of China, Iraq, or Libya? How about Microsoft or Doubleclick?
Unfortunately for Lars and Metallica, they want something ("artist control of their own music") that artists have never had, do not have under current copyright law, and for good public policy reasons ought never to have. Free-speech is a red herring--it's not the issue here; the right to free speech is completely distinct from the (fictional) right to be paid for your speech. Metallica has excercised their right to free speech by recording their music and having it published. Period, end of that story.
What's actually at issue is the real right (backed by copyright law) of the public, once having purchased Metallica's music, to do whatever they want with it, including making copies and sharing them with their friends. Free use of information, including things like recordings of Metallica performances, is legally recognized as the default state. As a matter of public policy in order to encourage the creation of artistic works, and not because of a "right" of an artist to be in permanent control of his work even after he's sold it, the law grants a limited, temporary monopoly on the commercial distribution of the artist's work--but that's the exception, not the rule--and the law makes it clear that the limited temporary monopoly does not override the underlying right of the consumer to use the work (including making copies for non-commercial purposes, excerpting for reviews, and all the other things that fall under "fair use").
The record companies have been doing everything in their power to convert that limited temporary monopoly into a permanent, unlimited one, and erase the fundamental distinction between information (which can be copied infinitely without making anyone poorer) and real property (which cannot).
Slashdot Mirror
It seems to me that's a solution looking for a problem. Don't like moderation? Read at -1. It's hardly a tyranny that the posts you most want to read aren't always at the top of the list. The typical (non-MS v. DOJ) story at Slashdot never has so many posts that it would be impractical to read them all if you're worried about missing a gem that wasn't moderated upward because it was posted too late or went against conventional wisdom (personally, I don't see the latter--you can generally find both sides of even contentious issues modded up). This is especially true since most of the lower ranked posts seem to be a single sentence.
If you stop worrying about the popularity contest of moderation, there are likely better ways to find interesting comments than an elaborate system of granting trust to editors. For instance, how about a button next to each post to grant a +/- 1 to other posts by the same author in your view. Using editors as a proxy for your tastes seems oddly indirect for something that you can filter yourself.
Part of the quid-pro-quo of the patent system (temporary monopoly in exchange for juicy details of invention) is that there has to be details of the invention that are non-trivial to implement/discover. If there aren't, then the public is giving up something naturally theirs (the right to use any idea) in exchange for nothing.
This is why according to law, the invention has to not only be original, but also must not be obvious to someone "skilled in the art"--i.e. who knows the technology in the area the patent covers. If it can be trivially reproduced (e.g. from a one-sentence description of what it does), then it was obvious.
The Amazon "one-click" patent ought not to have been awarded, even if it turns out that it was original, because once it is described, any web developer worth her salt can produce the equivalent in a few minutes work.
Because of the "obvious" nature of the idea, there's no public policy reason to grant Amazon a monopoly.
Patents (at least in the US) are not awarded as a gold star for having come up with an original idea, they are awarded to further the public good.
The framers of the Constitution recognized that nobody has or ought to have ownership of an idea, but in order to encourage people to share ideas instead of keeping them secret they made the patent system where you were granted a limited temporary monopoly on the use of your idea--in exchange for which you had to publish your idea, with all the detail necessary for someone else to implement it. Alternatively, you are perfectly within your rights to try to keep your idea secret, but if someone else comes up with the same idea, or your secret leaks out, you have no legal recourse.
So what about the one-click patent? Regardless of whether someone thought of it before, there's no public good in protecting an idea the very description of which gives anyone "skilled in the art" enough to go on to reproduce the idea. It is impossible for Amazon both to use this idea and to keep it a secret, therefor there's no reason to bargain with them to make it public.
You are wrong in your assumption.
The record company cuts her a check as an "advance on royalties"
Against this they charge her:
If, after all their creative accounting, there is any further "debt", they charge that against the next album as well. If the record makes money, the studio ends up not actually paying anything toward the production and distribution--it all comes out of the artist's share.
It's not only theoretically possible for the artist to owe the studio money after a profitable album, it happens all the time. Records go platinum, and the artist goes bankrupt.
So, yeah, Universal ought to owe all their artists a proportional chunk of that money, to show up on their royalty statements.
Presumably you also object to:
Nobody has any reasonable expectation that the price they pay will be the same from moment-to-moment regardless of the circumstances under which they buy. Nobody claims there's a conspiracy when two gas stations in the same franchise several blocks from eachother charge different prices, or different franchises across the street from one another. (In fact, quite the opposite: people charge price fixing when there's too much uniformity of price.) That's precisely the reason that some retailers offer a price guarantee as a selling point ("If we lower the price within the next 30 days, we'll refund the difference!").
If the price changes were separated by days, or were explicitly linked to things like which physical store they went into, or membership in a particular organization, nobody would even blink. It's only because the changes were happening at online speeds with no particular promotional effort that anyone thinks otherwise.
Suppose you're on top of the Empire State Building and want to get down to the sidewalk. You could either jump off (faster), or take the elevator down. They both have only two outcomes: You live or you die. Which do you choose?
If you chose the elevator, good for you: you understand why having two outcomes is not the same as having a 50-50 chance.
If you chose to jump, well, to quote Niven and Pournelle, "Think of it as evolution in action..."
A) You're assuming that Shasta and Pepsi are otherwise the same product, so that the only reason there is a difference in sales is advertising. Coca-Cola decided to put that theory to the acid test when they introduced "New Coke", reasoning that if they had a product that tasted just like Pepsi, they could slaughter them with advertising dollars. Didn't work so well, did it?
B) You're assuming that banner ads are just as effective as all other forms of advertising. That's hotly disputed.
C) As far as Double-Click making money--Double-Click is selling ads, not products. It's irrelevant whether Double-Click ads really work as long as they can get enough suckers who think that they work. And the dot-com economy proves there's no shortage of suckers willing to throw money at e-commerce.
Here's a pop-quiz for you: Name an e-commerce company that's making money selling to consumers (no infrastructure companies like Cisco). Now, if you've come up with one, describe one of their banner ads.
Have you ever noticed that whenever someone ends a sentence with ", thank you" (as in "I don't need an electronic conscience in my mail/new reader. My own will do just fine, thank you."), they never mean it? Same with sentences that begin with "I'll thank you to..."
Maybe someone could write a program that would filter for that...
I'd be a lot more concerned about all these attempts to capture "eyeballs" if there was any evidence at all that the eyeballs actually resulted in enough sales to make it worth the advertiser's while. As it is, I fully expect all of these schemes to collapse under the weight of massive consumer indifference.
Mozilla sent them, but Slashdot ignored them. If you want to control the formatting of your text on Slashdot, you need to learn to use the allowed HTML tags.
According to your thesis, this is your fault for not learning the nitty-gritty of the Slashdot interface.
I have nothing against the GPL, but consider what would have to happen before this bone of contention (re: whether the choice of law provision of CNRI's license is GPL compatible or not)would even be an issue.
Suppose you were to use Python 1.6+ under the CNRI license. Under what conditions could you be sued for mixing it with GPL code? As near as I can figure, you would have to have taken the Python code-base, extended it or embedded it within an application that you distribute AND have embedded someone else's GPL'ed code within your application. Then and only then you might be running the risk that the copyright holder of that other piece of code could try to sue you for having violated the terms of the GPL by including his code in a project that wasn't completely GPL compatible. (And if he did two out of the three sets of lawyers that have looked at both licenses agree that he wouldn't have a case.)
To me, this is a ridiculous thing to worry about...you're more likely to be struck by lightning or killed by terrorists than to become the test case where the alleged incompatibility of the GPL and the CNRI license is fought out in court.
Basically, Python tries to operate on the principle of least surprise. If you tell it to replace all the local function with the os versions of them, and then invoke the os open function, you shouldn't be too surprised if it behaves the way the native os open function does. You were surprised (this time), but you probably won't run into that again; consider how surprising it would be if it worked the other way around, though: despite explicitly telling the program to always prefer the os versions if available, the program persisted in secretly invoking the builtin version instead?
When you import another module, you do have access to that modules variables--they just don't pollute your local namespace unless you explicitly command them to (with "from spam import eggs" version of the import command). All you have to do is refer to them by their proper name: spam.eggs instead of just eggs.
If you really want to use eggs instead you can:
(sucks in all the names from spam)
eggs = spam.eggs
I don't know what you mean by "the writing function doesn't allow you to write strings." You can write pretty much anything you want, including binary.
Python has if/elif/else, for item in sequence, and while expression as forms of flow control. What else do you need? Endif and Wend are superfluous--they just signal the end of the block (Perl and C use }). Python can tell by the level of indentation.
GOTO. Don't need it. Between break, continue, and exception handling all the legitimate uses of GOTO as a flow control are covered (I suppose labelled breaks might be nice, but I haven't really missed them).
I don't understand your complaint about importing other modules as a form of flow control at all. I think you must have just missed the whole idea of functions and classes. What version of BASIC were you using before? Even VB doesn't use simple chaining of source-files for flow control--at least, not anymore.
The comp.lang.python newsgroup is bar far the friendliest, most informative, most newbie-tolerant newsgroup I've encountered on USENET. If that be hate-based, let's have more of it.
Also there's a report from XML.com on an paper presented at XML'99
There's an active XML SIG, with a page on Python org.
Actually, it's a long standing joke in comp.lang.python that Guido van Rossum, the creator of Python, has a time machine. It usually come up in the context of someone saying, "Boy, I really wish Python would add feature X." Then someone (usually Python guru Tim Peters) would respond to the effect that Guido likes the idea of adding feature X so much that he will use his time machine to retroactively add it to the language as of three releases ago....
"I am curious how one would access the same data in a Python construct - ie, how would a python programmer retrieve *all elements of an array pointed to by a an associative array reference* ? "
That's easy. Everything in Python is really a reference to an object, so in Python terms you're talking about an entry in a dictionary (Pythonese for an associative array) that happens itself to be an array object. Ergo:
foo["bar"]
More rights=good. Fewer rights=bad.
The argument against internet regulation remains the same. If installing extra hardware on every network would provide real security, then I think Bruce Schneier would endorse it. Rather, I believe the point not that we are not yet capable of a secure technology (whether implemented in hardware or software), but that there is no conceivable technology that is truly secure. The best you can do is realistically assess risks and mitigate them where it's cost-effective to do so.
As a Linux user, I'm certainly never going to install, let alone use, an integrated application that's designed with such a cavalier attitude towards security. Non-local communication should be the exception, not the norm, and should never occur invisibly.
...in that when you open a word processing document, you don't usually think that you're "doing something on the Internet." You especially don't think that you are communicating and setting cookies via invisible images.
"Thank you for your recent message to customer support. Please be advised that there will be a seven to ten day delay while our legal department reviews our customer support staff's reply. Thank you for your patience while we reduce our legal liability."
Find a sympathetic tenured professor and have him/her host whatever it was that attracted the MPAA letter. Academic institutions have completely different standards for freedom of speech for the faculty than the students. If nothing else, it should cause an entertaining ruckus.
Aside from the whole unproductive "How can you care about X when Y is so much more important?" whine (That's what it means to be free; deal with it.), you ignore the fact that modern technology--software above all--is the greatest threat to freedom of any kind that the world has ever seen. We are now capable of monitoring populations to a level of detail that exceed the wildest dreams of history's most oppresive regimes. We are rapidly approaching a point where surveillance can go from being total against a targeted individual to ubiquitous against all individuals. What price freedom then?
It is because freedom is more important than software that software is going to be one of the most important battlegrounds in future wars for freedom. Without free (as in speech and beer) software, that battle will be lost. (Heck, you won't even be able to tell that the battle has been engaged.)
Lest this sound like mere apocalyptic ranting, consider the FBI's Carnivore program. How much are you willing to trust the FBI that it will never be used against anyone but lawfully designated targets? Without full disclosure of the source, how much are you willing to trust that it can't be compromised by someone other than the FBI? How would your answer change if the authority providing the assurances were The People's Republic of China, Iraq, or Libya? How about Microsoft or Doubleclick?
What's actually at issue is the real right (backed by copyright law) of the public, once having purchased Metallica's music, to do whatever they want with it, including making copies and sharing them with their friends. Free use of information, including things like recordings of Metallica performances, is legally recognized as the default state. As a matter of public policy in order to encourage the creation of artistic works, and not because of a "right" of an artist to be in permanent control of his work even after he's sold it, the law grants a limited, temporary monopoly on the commercial distribution of the artist's work--but that's the exception, not the rule--and the law makes it clear that the limited temporary monopoly does not override the underlying right of the consumer to use the work (including making copies for non-commercial purposes, excerpting for reviews, and all the other things that fall under "fair use").
The record companies have been doing everything in their power to convert that limited temporary monopoly into a permanent, unlimited one, and erase the fundamental distinction between information (which can be copied infinitely without making anyone poorer) and real property (which cannot).